Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Three-Dimensional Diffusion-Weighted Multi-Slab MRI With Slice Profile Compensation Using Deep Energy Model

Reza Ghorbani, Jyothi Rikhab Chand, Chu-Yu Lee, Mathews Jacob, Merry Mani

TL;DR

This work tackles slab boundary artifacts in 3D diffusion-weighted multi-slab MRI by formulating a regularized slab profile encoding (PEN) problem solved within a Plug-and-Play ADMM framework. A Multi-Scale Energy (MuSE) regularizer is incorporated through a CNN-based energy model to robustly denoise and fuse slab-weighted images, enabling high-SNR, high-resolution reconstruction from under-sampled data. Compared to non-regularized PEN and TV-regularized PEN, MuSE-regularized PEN improves image quality and diffusion tensor metrics while reducing oversmoothing, though boundary artifacts persist due to RF pulse imperfections. The method holds promise for clearer anatomical imaging in clinical and research diffusion MRI, with future gains possible from improved RF pulse design and calibration.

Abstract

Three-dimensional (3D) multi-slab acquisition is a technique frequently employed in high-resolution diffusion-weighted MRI in order to achieve the best signal-to-noise ratio (SNR) efficiency. However, this technique is limited by slab boundary artifacts that cause intensity fluctuations and aliasing between slabs which reduces the accuracy of anatomical imaging. Addressing this issue is crucial for advancing diffusion MRI quality and making high-resolution imaging more feasible for clinical and research applications. In this work, we propose a regularized slab profile encoding (PEN) method within a Plug-and-Play ADMM framework, incorporating multi-scale energy (MuSE) regularization to effectively improve the slab combined reconstruction. Experimental results demonstrate that the proposed method significantly improves image quality compared to non-regularized and TV-regularized PEN approaches. The regularized PEN framework provides a more robust and efficient solution for high-resolution 3D diffusion MRI, potentially enabling clearer, more reliable anatomical imaging across various applications.

Three-Dimensional Diffusion-Weighted Multi-Slab MRI With Slice Profile Compensation Using Deep Energy Model

TL;DR

This work tackles slab boundary artifacts in 3D diffusion-weighted multi-slab MRI by formulating a regularized slab profile encoding (PEN) problem solved within a Plug-and-Play ADMM framework. A Multi-Scale Energy (MuSE) regularizer is incorporated through a CNN-based energy model to robustly denoise and fuse slab-weighted images, enabling high-SNR, high-resolution reconstruction from under-sampled data. Compared to non-regularized PEN and TV-regularized PEN, MuSE-regularized PEN improves image quality and diffusion tensor metrics while reducing oversmoothing, though boundary artifacts persist due to RF pulse imperfections. The method holds promise for clearer anatomical imaging in clinical and research diffusion MRI, with future gains possible from improved RF pulse design and calibration.

Abstract

Three-dimensional (3D) multi-slab acquisition is a technique frequently employed in high-resolution diffusion-weighted MRI in order to achieve the best signal-to-noise ratio (SNR) efficiency. However, this technique is limited by slab boundary artifacts that cause intensity fluctuations and aliasing between slabs which reduces the accuracy of anatomical imaging. Addressing this issue is crucial for advancing diffusion MRI quality and making high-resolution imaging more feasible for clinical and research applications. In this work, we propose a regularized slab profile encoding (PEN) method within a Plug-and-Play ADMM framework, incorporating multi-scale energy (MuSE) regularization to effectively improve the slab combined reconstruction. Experimental results demonstrate that the proposed method significantly improves image quality compared to non-regularized and TV-regularized PEN approaches. The regularized PEN framework provides a more robust and efficient solution for high-resolution 3D diffusion MRI, potentially enabling clearer, more reliable anatomical imaging across various applications.

Paper Structure

This paper contains 11 sections, 12 equations, 4 figures.

Table of Contents

  1. Introduction
  2. Methods
  3. Problem Formulation
  4. Energy Model in Multiple Dimensions
  5. MAP estimate using ADMM
  6. Implementation details
  7. Dataset
  8. Architecture and Training of Neural Network
  9. Experiments and results
  10. Conclusion
  11. Compliance with ethical standards

Figures (4)

  • Figure 1: Slab boundary artifact illustration.
  • Figure 2: T2 weighted images reconstructed using non-regularized PEN (a), TV-regularized PEN (b), and MuSE-regularized PEN (c).
  • Figure 3: Diffusion weighted images reconstructed using non-regularized PEN (a), TV-regularized PEN (b), and MuSE-regularized PEN (c).
  • Figure 4: FA (left), color-coded FA (middle), and MD (right) maps reconstructed using PEN(a), TV-regularized PEN (b), and MuSE-regularized PEN (c).